Hoffman Amplifiers Tube Amplifier Forum
Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: jordan86 on July 04, 2021, 05:16:40 pm
-
My brother is in town and I am working on his blues Junior. Reverb was not working. I pulled the tank and realized there was a connection inside the tank that was broken. So got that fixed.
What I have not been able to figure out is the unbalanced bias. I adjusted the bias so that it wasn’t crazy hot, like blues Junior‘s normally are. But what I am not sure how to fix is that the bias balance. I’ve tried four different sets of tubes and checked the bias. Even tried swapping tubes across sockets. It’s always the same side that is biasing about 15ma hotter. Makes me think mismatched tubes are not the problem. I’m sure this is common, I’m just not sure how to fix it. Any thoughts on what components or parts of the circuit to check/swap?
-
Leaky grid cap.
-
Leaky grid cap.
So I can measure for dcv present on C9 and C8?
https://www.thetubestore.com/lib/thetubestore/schematics/Fender/Fender-Blues-Junior-Schematic.pdf
-
Yes, after the cap. Measure both tube grids, they should be the same or very, very close.
The power tube that is measuring 15mA hotter, will be more positive than the cooler running tube. If the cap is leaking +dcv then it will cancel out the -bias -dcv.
If 1 coupling cap is bad, I'd change them both.
-
Check the voltage on pin 2, 7, and 9 of both EL84s. Looking for differences.
-
Thanks guys. Looking at my board, I think my schematic is on page 2 of that schematic PDF link. Page 1 is older model I think. C15 and C16 make way more sense on my board.
-
figure out which wire is the grid, follow the foil.
check all the gray cables, the solder inspector was flirting on every 3rd batch :icon_biggrin:
-
Well....Took some voltage readings on the power tubes. They were disappointingly similar. FWIW It is the brown primary wire side that always runs hotter.
V4 (Blue primary side)
Pin 2: -8.73 vdc
Pin 7: 329 vdc
Pin 9: 303 vdc
V5 (Brown primary side)
Pin 2: -8.72 vdc
Pin 7: 327 vdc
Pin 9: 303 vdc
I was expecting Pin 2 to be different based on the leaky cap theory. But no. Also I was not able to check for dc voltage directly off the coupling caps feeding into the EL84s grids (C15/C16). Theres no lead exposed above the pcb to get a probe on.
The grid stoppers and grid leak resistors on V4/V5 all measure in spec and match on each side too.
One other interesting thing is that I do get a notable hum when I probe the blue primary side plate. Pretty loud. And my multimeter kinda wigs out. Instead of displaying 330V, or -2.41v when measure the drop on CT, the meter will read: .3.3.0.v or -.2.4.1.v. Maybe it’s trying to tell me something. Guess I need to check the manual.
-
One other interesting thing is that I do get a notable hum when I probe the blue primary side plate. Pretty loud. And my multimeter kinda wigs out. Instead of displaying 330V, or -2.41v when measure the drop on CT, the meter will read: .3.3.0.v or -.2.4.1.v. Maybe it’s trying to tell me something. Guess I need to check the manual.
I've had a similar multimeter wig out when measuring a friends blues junior. It didn't like my cheap meter or my scope, was ok with my fluke. You're basically adding a big loop (aerial) of wire and electronics to a high voltage node in the circuit with unpredictable results. Ok on some amps, mabe not with the blues junior.
-
Pin 2: -8.73 vdc
This seems too low to me. Probably running those tubes very hot. All my P/P EL84 amps are running about -12v for proper bias.
I'm assuming this amp has been modified to have a bias pot? Describe your method of measuring bias. You say one tube always runs 15mA hotter, but you never said what the actual currents were. What are they?
And what brand/model number is your meter?
-
I have a cheapo Craftsman MM. My neighbor has a Fluke though. I can borrow his today to see if that changes anything.
As for bias checking, I am measuring the voltage drop across the OT primaries to the Center tap. Then power down and measuring resistance across those same spots. For ballpark, I am usually getting a 3.5-3.8v drop on the hot side, and roughly 2-2.4v drop on the cool side. About 99 ohm resistance. And ~330v on the plates.
Strangely...After my post last night, I paralleled a 100K resistor with R51 the bias circuit, for a bias voltage of -12.68v. Stock is -10.6v. Tried another set of tubes. Gave me 9w and 7w plate dissipation respectively. Much closer. Did not measure grid voltage at Pin 2. But that’s almost close enough for me to call it done and quit caring.
If you have any other ideas though, I’d be all ears.
-
Pin 2: -8.73 vdc
This seems too low to me. Probably running those tubes very hot. All my P/P EL84 amps are running about -12v for proper bias.
I'm assuming this amp has been modified to have a bias pot?
No bias pot Sluckey. Too lazy to pull the pcb and drill. Been paralleling resistors in the bias circuit and experimenting. It that hot though, too hot. Normal for a blues jr as I understand it. They come from the factory wicked hot for some reason. I’m hoping to get the bias matched, cooler with a tackle solder resistor from above, and install some of my 3,000+ hour 6n14n’s. Should last my brother a lifetime with how much he plays.
-
I am measuring the voltage drop across the OT primaries to the Center tap.
That plus a cheap MM probably totally accounts for your difference in current numbers. Based on the tube voltages you posted I suspect there really is no difference between the real current flowing through the tubes.
You can get a much more reliable current reading by installing a 1Ω 1% 1W resistor between pin 3 and ground for each EL84. Probably require cutting the pcb trace to pin 3.
You still have not said what the tube currents are. You've only said one is 15mA hotter. Asking once more, what are your actual calculated tube currents?
-
You can get a much more reliable current reading by installing a 1Ω 1% 1W resistor between pin 3 and ground for each EL84. Probably require cutting the pcb trace to pin 3.
Yes, I know there are better methods, but not sure I'm up for major surgery.
You still have not said what the tube currents are. You've only said one is 15mA hotter. Asking once more, what are your actual calculated tube currents?
In the stock configuration (no bias adjustment), I am calculating ~38ma on the hot side and ~24ma on the cold. This is with the transformer voltage Drop divided by resistance method, times plate voltage. With the 100K in the bias circuit, I got 27ma and 21ma.
-
I have several Blues Juniors, all modded about 5 years ago. My recollection is that 21 vs 27 mA is what you normally see in this amp. And, Fender biased these pretty hot from the factory.
One mod I've tried is use a voltage divider (and trimmer pot) to allow balancing the EL-84's better. Frankly, I heard little difference in tone, though perhaps cleaner when turned up louder than my old ears can take.
However, biasing the tubes colder does result in much improved life of the EL-84s. It was worth it to me, as I played daily, and it was a hassle, albeit a minor one, when the tubes would fail every few months.
-
I believe your method is faulty and the circuit is fine. Do you have access to a bias probe? If not, may be time to build or buy one. On second thought, you probably can't get the probe sockets to plug into the board mounted sockets.
Still easier and quicker and cheaper to just do the resistor mod. Just get access to the trace side of the pcb, dremel or razor blade utility knife to cut the trace, solder resistor across the trace gap you just cut. This is not major surgery.
-
FWIW I got a Mercury Magnetics “Tone Clone” OT for my Blues Jr. years ago. The resistance was seriously mismatched between the 2 OT secondaries. When I called MM to say I had a defective OT, they insisted it was a perfect clone of a Fender Blues Jr. OT.
My Blues Jr. is a 2001 green board version and the OT secondaries have almost identical resistance.
Don’t know if that helps.
BTW I think you are wise to do as little as possible inside that amp. I put in adjustable bias, but those ribbon cables, etc. are so fragile that just moving the PCB caused other problems.
Cheers,
Chip
-
The tube circuit board is connected to the main PCB by the snaky gray cables. Pin 3 returns to the mainboard ground via PW4B-4 and PW5B-1 (this is true for all variants, AFAIK).
It would be fairly simple to dissect out those wires and place a 1 ohm resistor in the path. Maybe a bit of a kludge, but NOT major surgery.
-
I recently repaired a BJ that had a catastrophic meltdown the took out the ot. The symptom after tube replacement was the bias was very unbalanced. The oversized Fromel MM OT that was in it had way unbalanced primary windings. Ordered a BJ OT pull from Fromel. When it arrived I tested it and it had very balance primary windings.
-
I suggest not to probe the anodes of output valves. It can induce oscillation, and the voltage swing there can be massive, far greater than any regular meter’s rating.
For the anode voltage, just measure the HT voltage supplied to the OT.
If you’re using the OT volt drop method to check anode idle current / dissipation, then it’s a very good idea to remove the valve in the phase splitter socket when doing so.
-
My OT has pretty balanced windings. When measuring resistance, they measure within less than .5 ohms difference. Gonna borrow my neighbors fluke and do some more investigating.
-
BTW I think you are wise to do as little as possible inside that amp. I put in adjustable bias, but those ribbon cables, etc. are so fragile that just moving the PCB caused other problems.
Yes, as least invasive as possible is my goal. I did watch BillM's video on puling the circuit board. Seems simple enough, but it's nearly a 20 year old amp, and I'd rather not play roulette. If something is majorly wrong, I'll prob order some new filter caps, drop the board and replace those while I'm in, but I consider that a last resort.
-
If you’re using the OT volt drop method to check anode idle current / dissipation, then it’s a very good idea to remove the valve in the phase splitter socket when doing so.
Well, Sluckey and PDF64 were right. I pulled the phase inverter 12AX7. Repeated the process with the Fluke, the tubes measure 28.6ma and 27.95ma. Pin 2 readings match. And the voltage drop across the OT is very close as well. I call that matched.
Strangely though, when I put the PI tube back in, I get that hum when probing the blue side OT/plate and my fluke reads OL. Would that point to some other issue with the PI, or is that "normal" (expected)? I tried different PI tubes. No change.
-
It's all good.
-
> I get that hum when probing the blue side OT/plate
Stop doing that!!
There is a LOT of gain from PI input to EL84 plate. If you hang a 2 foot wire on that plate, it radiates a lot of signal, some back to the PI or preamps. It squeals, just like a mike and a speaker in the same room. The squeal may be supersonic. The hum is the increased load and generally crappy audio performance when squealed-out.
I am also dubious about matching two sides of an OT. It is not important, and the common way to wind (one over the other) will make 20%-30% difference. In DC resistance. Which is far less than Audio impedance. 50 Ohms difference in a 8000 ohm winding, is nothing.